On Sat, 29 Nov 1997, Johan Klockars wrote:
> > >From memory:
> > 14 bytes are shifted +16 bytes in position in the file but I think 2
> > bytes are somehow skipped.
> > Then alternate runs of 16 bytes are exchanged.
> > At some point an extra 2 bytes is inserted and the corruption ends.
>
> I asked Uwe Seimet about those rotations, thinking that there might be some
> 16 byte buffer in the SCSI DMA hardware that was having problems.
> According to him, however, there is no such thing on the Falcon...
a cache line is 16 bytes long, perhaps?
> I'd really like to know how those rotations take place. That might give some
> clue as to what's going on.
try to ask in Linux-m68k mailing list, they spent ages on solving such
errors, too (on a TT, however).
Petr
--
WWW: http://ft3.zlin.vutbr.cz/stehlik/home.htm PARCP development
e-mail: stehlik_at_cas3.zlin.vutbr.cz MiNT/Linux user
mirror: http://www.cyberstrider.org/mirrors/stehlik/ AfterBurner040 owner
Received on sø. nov. 30 1997 - 22:36:00 CET